The use of HIV protease inhibitors (PIS), combined with other anti-retroviral agents, is central to dramatic decline in the morbidity and mortality of HIV infection. Our group recently discovered that PI-treated patients have decreased bone mineral density (BMD), which led us to examine the effects of individual PIS on osteoblast (OB) and osteoclast (OC) formation and function. We find that one PI, Indinavir, induces bone loss in mice. The drug exerts its osteoporotic effect by arresting OB precursor differentiation, and in so doing, dampens expression of the specific osteoblast transcription factor, Cbfal. In contrast to the osteoporotic properties of Indinavir, Ritonavir is bone sparing, in vivo, exerting its effect, in an M-CSF independent manner, by blunting OC formation and function. Since osteoclastogenesis requires only M-CSF and RANK ligand (RANKL) the drug must impact RANKL-initiated signaling. Finally, we provide data that a fusion protein comprising GST linked to the ectodomain of RANKL is a potent bone anabolic agent. We therefore hypothesize that: 1. lndinavir arrests bone formation by inhibiting Cbfal expression in OB precursors; 2. Ritonavir, by blunting RANKL-induced signaling, inhibits OC differentiation and function; and 3. Administration of Ritonavir, or GST-RANKL, will prevent andlor rescue Indinavir-induced osteoporosis. Thus, our specific aims are to determine: 1. the mechanism by which lndinavir inhibits Cbfal expression in OB precursors; 2. The mechanism by which Ritonavir, blunts RANKL-induced signaling, in OCs and their precursors; 3. If administration of Ritonavir, or GST-RANKL, will prevent and/or rescue Indinavir-induced osteoporosis.